Treatment of HIV-infected individuals with protease inhibitors (PIs) is associated with changes in body fat distribution and metabolic disturbances similar to those seen in Syndrome X and the partial lipodystrophies. PI use has been variably associated with hypertriglyceridemia, hypercholesterolemia, insulin resistance including diabetes mellitus and the development of central obesity, but it is unclear how these complications relate to each other. Based on the experience of patients with Syndrome X, it is highly probable that if these metabolic complications of PI therapy cluster, then they will in turn be associated with an increased risk of coronary artery disease. A major hypothesis of this proposal is that patients who develop an increase in abdominal girth or hypertriglyceridemia without a clinically apparent change in body fat distribution will have both significantly more visceral fat and insulin resistance compared to controls. To teat this hypothesis, four groups of HIV-infected subjects will be compared: PI-treated patients who have developed a "protease paunch", PI-treated patients with hyper- triglyceridemia but no clinically apparent change in body fat distribution, PI-treated patients free of these two complications and PI-naove patients. Body composition and visceral adiposity will be determined by DEXA and abdominal CT respectively. Insulin sensitivity will be measured using the FISGTT. In addition, lipoprotein electrophoresis and FPLC will be used to characterize lipoprotein profiles.